!pip install geopandas
!pip install contextily
!pip install seaborn as sns
!pip install folium
!pip install sodapy
!pip install cartopy

import zipfile
import geopandas as gpd
import contextily as ctx
from shapely.geometry import Point
import pandas as pd
import folium
import numpy as np
import matplotlib.pyplot as plt
import datetime
import matplotlib.dates as mdates
import seaborn as sns
import plotly.express as px
import pandas as pd
from sodapy import Socrata
import string
import cartopy
import cartopy.crs as crs

Requirement already satisfied: geopandas in /opt/conda/lib/python3.8/site-packages (0.12.1)
Requirement already satisfied: pandas>=1.0.0 in /opt/conda/lib/python3.8/site-packages (from geopandas) (1.1.2)
Requirement already satisfied: packaging in /opt/conda/lib/python3.8/site-packages (from geopandas) (22.0)
Requirement already satisfied: pyproj>=2.6.1.post1 in /opt/conda/lib/python3.8/site-packages (from geopandas) (2.6.1.post1)
Requirement already satisfied: fiona>=1.8 in /opt/conda/lib/python3.8/site-packages (from geopandas) (1.8.18)
Requirement already satisfied: shapely>=1.7 in /opt/conda/lib/python3.8/site-packages (from geopandas) (1.7.1)
Requirement already satisfied: pytz>=2017.2 in /opt/conda/lib/python3.8/site-packages (from pandas>=1.0.0->geopandas) (2020.1)
Requirement already satisfied: numpy>=1.15.4 in /opt/conda/lib/python3.8/site-packages (from pandas>=1.0.0->geopandas) (1.19.1)
Requirement already satisfied: python-dateutil>=2.7.3 in /opt/conda/lib/python3.8/site-packages (from pandas>=1.0.0->geopandas) (2.8.1)
Requirement already satisfied: attrs>=17 in /opt/conda/lib/python3.8/site-packages (from fiona>=1.8->geopandas) (20.2.0)
Requirement already satisfied: certifi in /opt/conda/lib/python3.8/site-packages (from fiona>=1.8->geopandas) (2022.9.24)
Requirement already satisfied: click<8,>=4.0 in /opt/conda/lib/python3.8/site-packages (from fiona>=1.8->geopandas) (7.1.2)
Requirement already satisfied: cligj>=0.5 in /opt/conda/lib/python3.8/site-packages (from fiona>=1.8->geopandas) (0.7.2)
Requirement already satisfied: click-plugins>=1.0 in /opt/conda/lib/python3.8/site-packages (from fiona>=1.8->geopandas) (1.1.1)
Requirement already satisfied: six>=1.7 in /opt/conda/lib/python3.8/site-packages (from fiona>=1.8->geopandas) (1.15.0)
Requirement already satisfied: munch in /opt/conda/lib/python3.8/site-packages (from fiona>=1.8->geopandas) (2.5.0)
Requirement already satisfied: contextily in /opt/conda/lib/python3.8/site-packages (1.2.0)
Requirement already satisfied: mercantile in /opt/conda/lib/python3.8/site-packages (from contextily) (1.2.1)
Requirement already satisfied: joblib in /opt/conda/lib/python3.8/site-packages (from contextily) (0.17.0)
Requirement already satisfied: xyzservices in /opt/conda/lib/python3.8/site-packages (from contextily) (2022.9.0)
Requirement already satisfied: matplotlib in /opt/conda/lib/python3.8/site-packages (from contextily) (3.3.2)
Requirement already satisfied: geopy in /opt/conda/lib/python3.8/site-packages (from contextily) (2.3.0)
Requirement already satisfied: rasterio in /opt/conda/lib/python3.8/site-packages (from contextily) (1.2.1)
Requirement already satisfied: pillow in /opt/conda/lib/python3.8/site-packages (from contextily) (7.2.0)
Requirement already satisfied: requests in /opt/conda/lib/python3.8/site-packages (from contextily) (2.28.1)
Requirement already satisfied: click>=3.0 in /opt/conda/lib/python3.8/site-packages (from mercantile->contextily) (7.1.2)
Requirement already satisfied: numpy>=1.15 in /opt/conda/lib/python3.8/site-packages (from matplotlib->contextily) (1.19.1)
Requirement already satisfied: kiwisolver>=1.0.1 in /opt/conda/lib/python3.8/site-packages (from matplotlib->contextily) (1.2.0)
Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.3 in /opt/conda/lib/python3.8/site-packages (from matplotlib->contextily) (2.4.7)
Requirement already satisfied: python-dateutil>=2.1 in /opt/conda/lib/python3.8/site-packages (from matplotlib->contextily) (2.8.1)
Requirement already satisfied: certifi>=2020.06.20 in /opt/conda/lib/python3.8/site-packages (from matplotlib->contextily) (2022.9.24)
Requirement already satisfied: cycler>=0.10 in /opt/conda/lib/python3.8/site-packages (from matplotlib->contextily) (0.10.0)
Requirement already satisfied: geographiclib<3,>=1.52 in /opt/conda/lib/python3.8/site-packages (from geopy->contextily) (1.52)
Requirement already satisfied: snuggs>=1.4.1 in /opt/conda/lib/python3.8/site-packages (from rasterio->contextily) (1.4.7)
Requirement already satisfied: affine in /opt/conda/lib/python3.8/site-packages (from rasterio->contextily) (2.3.1)
Requirement already satisfied: cligj>=0.5 in /opt/conda/lib/python3.8/site-packages (from rasterio->contextily) (0.7.2)
Requirement already satisfied: attrs in /opt/conda/lib/python3.8/site-packages (from rasterio->contextily) (20.2.0)
Requirement already satisfied: click-plugins in /opt/conda/lib/python3.8/site-packages (from rasterio->contextily) (1.1.1)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in /opt/conda/lib/python3.8/site-packages (from requests->contextily) (1.25.10)
Requirement already satisfied: idna<4,>=2.5 in /opt/conda/lib/python3.8/site-packages (from requests->contextily) (2.10)
Requirement already satisfied: charset-normalizer<3,>=2 in /opt/conda/lib/python3.8/site-packages (from requests->contextily) (2.1.1)
Requirement already satisfied: six>=1.5 in /opt/conda/lib/python3.8/site-packages (from python-dateutil>=2.1->matplotlib->contextily) (1.15.0)
Requirement already satisfied: seaborn in /opt/conda/lib/python3.8/site-packages (0.11.0)
ERROR: Could not find a version that satisfies the requirement as (from versions: none)
ERROR: No matching distribution found for as
Requirement already satisfied: folium in /opt/conda/lib/python3.8/site-packages (0.13.0)
Requirement already satisfied: numpy in /opt/conda/lib/python3.8/site-packages (from folium) (1.19.1)
Requirement already satisfied: branca>=0.3.0 in /opt/conda/lib/python3.8/site-packages (from folium) (0.6.0)
Requirement already satisfied: jinja2>=2.9 in /opt/conda/lib/python3.8/site-packages (from folium) (2.11.2)
Requirement already satisfied: requests in /opt/conda/lib/python3.8/site-packages (from folium) (2.28.1)
Requirement already satisfied: MarkupSafe>=0.23 in /opt/conda/lib/python3.8/site-packages (from jinja2>=2.9->folium) (1.1.1)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in /opt/conda/lib/python3.8/site-packages (from requests->folium) (1.25.10)
Requirement already satisfied: charset-normalizer<3,>=2 in /opt/conda/lib/python3.8/site-packages (from requests->folium) (2.1.1)
Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.8/site-packages (from requests->folium) (2022.9.24)
Requirement already satisfied: idna<4,>=2.5 in /opt/conda/lib/python3.8/site-packages (from requests->folium) (2.10)
Requirement already satisfied: sodapy in /opt/conda/lib/python3.8/site-packages (2.2.0)
Requirement already satisfied: requests>=2.28.1 in /opt/conda/lib/python3.8/site-packages (from sodapy) (2.28.1)
Requirement already satisfied: charset-normalizer<3,>=2 in /opt/conda/lib/python3.8/site-packages (from requests>=2.28.1->sodapy) (2.1.1)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in /opt/conda/lib/python3.8/site-packages (from requests>=2.28.1->sodapy) (1.25.10)
Requirement already satisfied: idna<4,>=2.5 in /opt/conda/lib/python3.8/site-packages (from requests>=2.28.1->sodapy) (2.10)
Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.8/site-packages (from requests>=2.28.1->sodapy) (2022.9.24)
Requirement already satisfied: cartopy in /opt/conda/lib/python3.8/site-packages (0.18.0)
Requirement already satisfied: setuptools>=0.7.2 in /opt/conda/lib/python3.8/site-packages (from cartopy) (49.6.0.post20200917)
Requirement already satisfied: shapely>=1.5.6 in /opt/conda/lib/python3.8/site-packages (from cartopy) (1.7.1)
Requirement already satisfied: pyshp>=1.1.4 in /opt/conda/lib/python3.8/site-packages (from cartopy) (2.3.1)
Requirement already satisfied: numpy>=1.10 in /opt/conda/lib/python3.8/site-packages (from cartopy) (1.19.1)
Requirement already satisfied: six>=1.3.0 in /opt/conda/lib/python3.8/site-packages (from cartopy) (1.15.0)

#!/usr/bin/env python

# Unauthenticated client only works with public data sets. Note 'None'
# in place of application token, and no username or password:
client = Socrata("data.sfgov.org", None)

results = client.get("wg3w-h783", limit=1000000)

# Convert to pandas DataFrame
df1 = pd.DataFrame.from_records(results)
df = df1

WARNING:root:Requests made without an app_token will be subject to strict throttling limits.

df.columns

Index(['incident_datetime', 'incident_date', 'incident_time', 'incident_year',
       'incident_day_of_week', 'report_datetime', 'row_id', 'incident_id',
       'incident_number', 'report_type_code', 'report_type_description',
       'filed_online', 'incident_code', 'incident_category',
       'incident_subcategory', 'incident_description', 'resolution',
       'police_district', 'cad_number', 'intersection', 'cnn',
       'analysis_neighborhood', 'supervisor_district', 'latitude', 'longitude',
       'point', ':@computed_region_jwn9_ihcz', ':@computed_region_26cr_cadq',
       ':@computed_region_qgnn_b9vv', ':@computed_region_nqbw_i6c3',
       ':@computed_region_h4ep_8xdi', ':@computed_region_n4xg_c4py',
       ':@computed_region_jg9y_a9du'],
      dtype='object')

df.drop(columns=[':@computed_region_jwn9_ihcz', ':@computed_region_26cr_cadq', ':@computed_region_qgnn_b9vv',
                ':@computed_region_nqbw_i6c3', ':@computed_region_h4ep_8xdi',':@computed_region_n4xg_c4py',
                ':@computed_region_jg9y_a9du', 'point'], inplace = True)

df['incident_datetime']=pd.to_datetime(df['incident_datetime'], errors = 'coerce')
df['incident_year']=pd.to_datetime(df['incident_year'], errors = 'coerce')
df['incident_date']=pd.to_datetime(df['incident_date'], errors = 'coerce')
df['report_datetime']=pd.to_datetime(df['report_datetime'], errors = 'coerce')

df.dtypes

incident_datetime          datetime64[ns]
incident_date              datetime64[ns]
incident_time                      object
incident_year              datetime64[ns]
incident_day_of_week               object
report_datetime            datetime64[ns]
row_id                             object
incident_id                        object
incident_number                    object
report_type_code                   object
report_type_description            object
filed_online                       object
incident_code                      object
incident_category                  object
incident_subcategory               object
incident_description               object
resolution                         object
police_district                    object
cad_number                         object
intersection                       object
cnn                                object
analysis_neighborhood              object
supervisor_district                object
latitude                           object
longitude                          object
dtype: object

def o_str(value):
    return str(value)

def o_date(value):
    return str(value)

def o_numeric(value):
    return float(value)

df['incident_id']=df['incident_id'].apply(o_numeric)
df['row_id']=df['row_id'].apply(o_numeric)
df['incident_code']=df['incident_code'].apply(o_numeric)
df['incident_number']=df['incident_number'].apply(o_numeric)
df['cad_number']=df['cad_number'].apply(o_numeric)
df['cnn']=df['cnn'].apply(o_numeric)

df['report_type_description']=df['report_type_description'].apply(o_str)
df['incident_category']=df['incident_category'].apply(o_str)
df['incident_subcategory']=df['incident_subcategory'].apply(o_str)
df['resolution']=df['resolution'].apply(o_str)
df['police_district']=df['police_district'].apply(o_str)
df['analysis_neighborhood']=df['analysis_neighborhood'].apply(o_str)

df['latitude']=df['latitude'].apply(o_numeric)
df['longitude']=df['longitude'].apply(o_numeric)

df.resolution.unique()

array(['Open or Active', 'Cite or Arrest Adult', 'Exceptional Adult',
       'Unfounded'], dtype=object)

df.police_district.unique()

array(['Southern', 'Out of SF', 'Central', 'Mission', 'Richmond',
       'Ingleside', 'Park', 'Northern', 'Bayview', 'Tenderloin',
       'Taraval'], dtype=object)

df.report_type_description.unique()

array(['Coplogic Initial', 'Vehicle Supplement', 'Initial Supplement',
       'Initial', 'Vehicle Initial', 'Coplogic Supplement'], dtype=object)

df.analysis_neighborhood.unique()

array(['nan', 'Excelsior', 'Russian Hill', 'Lone Mountain/USF',
       'Pacific Heights', 'Bayview Hunters Point', 'Glen Park', 'Mission',
       'Tenderloin', 'Hayes Valley', 'Western Addition', 'Outer Mission',
       'Marina', 'South of Market', 'Japantown', 'McLaren Park',
       'Presidio Heights', 'Mission Bay', 'Nob Hill', 'North Beach',
       'Financial District/South Beach', 'West of Twin Peaks',
       'Chinatown', 'Haight Ashbury', 'Castro/Upper Market',
       'Sunset/Parkside', 'Potrero Hill', 'Lakeshore', 'Outer Richmond',
       'Inner Sunset', 'Twin Peaks', 'Oceanview/Merced/Ingleside',
       'Portola', 'Presidio', 'Bernal Heights', 'Golden Gate Park',
       'Noe Valley', 'Visitacion Valley', 'Inner Richmond',
       'Treasure Island', 'Lincoln Park', 'Seacliff', 'null'],
      dtype=object)

# Filling in the NA values with In Person, as specified on the website, 
# any row which does not have Filled Online it is assumed that the complaint was filed in person

df['filed_online'].fillna('In Person',inplace = True)

df.dtypes

incident_datetime          datetime64[ns]
incident_date              datetime64[ns]
incident_time                      object
incident_year              datetime64[ns]
incident_day_of_week               object
report_datetime            datetime64[ns]
row_id                            float64
incident_id                       float64
incident_number                   float64
report_type_code                   object
report_type_description            object
filed_online                       object
incident_code                     float64
incident_category                  object
incident_subcategory               object
incident_description               object
resolution                         object
police_district                    object
cad_number                        float64
intersection                       object
cnn                               float64
analysis_neighborhood              object
supervisor_district                object
latitude                          float64
longitude                         float64
dtype: object

#Checking null values for each column
df.isnull().sum()

incident_datetime               0
incident_date                   0
incident_time                   0
incident_year                   0
incident_day_of_week            0
report_datetime                 0
row_id                          0
incident_id                     0
incident_number                 0
report_type_code                0
report_type_description         0
filed_online                    0
incident_code                   0
incident_category               0
incident_subcategory            0
incident_description            0
resolution                      0
police_district                 0
cad_number                 150519
intersection                35674
cnn                         35674
analysis_neighborhood           0
supervisor_district         35674
latitude                    35674
longitude                   35674
dtype: int64

#Since over 70% of the data is null drop this columns

df.cad_number.fillna('0', inplace = True)
df.intersection.fillna('0', inplace = True)
df.cnn.fillna('0', inplace = True)
df.supervisor_district.fillna('0', inplace = True)


df['latitude'].fillna(float(df['latitude'].mean()), inplace=True)
df['longitude'].fillna(float(df['longitude'].mean()), inplace=True)

df.analysis_neighborhood.fillna('0', inplace = True)

#Reindexing and finding shape of dataframe
df.iloc[:].reindex()
df.shape

(673198, 25)

df.rename(columns={'incident_category': 'Incident Category'}, inplace = True)

This graph shows us the Count for Number of Crimes occuring in the SFO region based on Days

# Barplot showing major crimes in San Francisco

plt.rcParams['figure.figsize'] = (20, 9)
plt.style.use('fast')
sns.countplot(df['Incident Category'], palette = 'hot')

plt.title('Major Crimes in San Francisco', fontweight = 20, fontsize = 20)
plt.xticks(rotation = 90)
plt.show()

/opt/conda/lib/python3.8/site-packages/seaborn/_decorators.py:36: FutureWarning: Pass the following variable as a keyword arg: x. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.
  warnings.warn(

Distribution of crimes in San Francisco as per days of week

# Distribution of crimes in San Francisco as per days of week

df['incident_day_of_week'].value_counts().plot.pie(figsize = (10, 20), explode = (0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1), autopct='%1.1f%%')
plt.axis('off')
plt.title('Crime count on each day',fontsize = 20)
centre_circle = plt.Circle((0,0),0.6,fc='white')
fig = plt.gcf()
fig.gca().add_artist(centre_circle)
plt.xticks(rotation = 90)
plt.show()

# Get a count of the crimes by neighborhood, returns a series
crime_neighbourhood = df.police_district.value_counts()
# Stick it into a data frame.
neighbour_df = pd.DataFrame(crime_neighbourhood)

# Set up the data frame with the required structure.
neighbour_df.index.name = 'Neighborhood'
neighbour_df.reset_index(inplace=True)
neighbour_df.rename(columns={'police_district': 'No of Crimes'}, inplace=True)
neighbour_df

	Neighborhood	No of Crimes
0	Central	100613
1	Northern	91518
2	Mission	84718
3	Southern	81150
4	Tenderloin	64532
5	Bayview	59492
6	Ingleside	50833
7	Taraval	47267
8	Richmond	41986
9	Park	31274
10	Out of SF	19815

Distribution of Number of Crimes that has occured over the period of last 4 years

x = df.groupby(['Incident Category', 'incident_year'])[['incident_year']].count()
x.rename(columns={"incident_year":"count"}, inplace=True)
crime_by_years = x.reset_index()
crime_by_years

	Incident Category	incident_year	count
0	Arson	2018-01-01	366
1	Arson	2019-01-01	313
2	Arson	2020-01-01	429
3	Arson	2021-01-01	433
4	Arson	2022-01-01	379
...	...	...	...
241	nan	2018-01-01	10
242	nan	2019-01-01	2
243	nan	2020-01-01	306
244	nan	2021-01-01	153
245	nan	2022-01-01	88

246 rows × 3 columns

df['Incident Category'] = df['Incident Category'].replace(['Motor Vehicle Theft?', 'Other Miscellaneous', 'Other Offenses', 'Weapons Offence'], 
                                                          ['Motor Vehicle Theft', 'Other', 'Other', 'Weapons Offense'])

df['incident_subcategory'] = df['incident_subcategory'].replace(['Motor Vehicle Theft?', 'Other Miscellaneous', 'Other Offenses', 'Weapons Offence'], 
                                                          ['Motor Vehicle Theft', 'Other', 'Other', 'Weapons Offense'])

fig = px.bar(crime_by_years, x='incident_year', y='count', color='Incident Category', 
            title="Counts of crimes according to categories grouped over the years",
            labels={'incident_year':'Year', 'count':'Count of crimes'})
fig.show('notebook')

Average delay time between the Incident happening and the Report time

df['time_difference'] = round(((df['report_datetime'] - df['incident_datetime']).dt.total_seconds() / 60 / 60 / 24),0)
df[['time_difference']]

	time_difference
0	1.0
1	0.0
2	0.0
3	0.0
4	0.0
...	...
673193	1.0
673194	0.0
673195	0.0
673196	0.0
673197	0.0

673198 rows × 1 columns

x = df.groupby(['incident_year'])['time_difference'].mean()
time_diff_by_year = x.reset_index()
time_diff_by_year

	incident_year	time_difference
0	2018-01-01	10.157525
1	2019-01-01	7.834363
2	2020-01-01	10.588955
3	2021-01-01	6.911124
4	2022-01-01	3.817991

fig = px.line(time_diff_by_year, x='incident_year', y='time_difference',
              title="Average Time difference between incident and report time in days over the years",
              labels={'incident_year':'Year', 'time_difference':'Time difference in days'})
fig.show()

Bargraph showing distribution of incidences as per the time of its occurance

df['incident_time']=df['incident_time'].astype('string')

df['incident_time']=df['incident_time'].str.split(':',expand=True)[0]
df['incident_time']=df['incident_time'].astype(int)

df['incident_time_category']='null'
df['incident_time_category'].dtypes

dtype('O')

# 00.00 - 06.00 : Early Morning, 6.01 - 12.00 : Morning, 12.01 - 18.00 : Evening, 18.01 - 24.00 : Night

def category(x):
    if 0<=x<=6:
        return 'Early Morning'
    elif 6<x<=12:
        return 'Morning'
    elif 12<x<=18:
        return 'Evening'
    elif 18<x<24:
        return 'Night'

df['incident_time_category'] = df['incident_time'].apply(category)

df_count = df.groupby(['incident_time_category']).count()
df_count

	incident_datetime	incident_date	incident_time	incident_year	incident_day_of_week	report_datetime	row_id	incident_id	incident_number	report_type_code	...	resolution	police_district	cad_number	intersection	cnn	analysis_neighborhood	supervisor_district	latitude	longitude	time_difference
incident_time_category
Early Morning	115357	115357	115357	115357	115357	115357	115357	115357	115357	115357	...	115357	115357	115357	115357	115357	115357	115357	115357	115357	115357
Evening	227333	227333	227333	227333	227333	227333	227333	227333	227333	227333	...	227333	227333	227333	227333	227333	227333	227333	227333	227333	227333
Morning	177706	177706	177706	177706	177706	177706	177706	177706	177706	177706	...	177706	177706	177706	177706	177706	177706	177706	177706	177706	177706
Night	152802	152802	152802	152802	152802	152802	152802	152802	152802	152802	...	152802	152802	152802	152802	152802	152802	152802	152802	152802	152802

4 rows × 26 columns

df_count = df.groupby(['incident_time_category']).count()
df_num=pd.DataFrame(df_count)
df_num.reset_index(inplace=True)

fig = plt.figure(figsize = (10, 7))
fig = px.bar(df_num.sort_values( 'incident_time',ascending = False), 
            x='incident_time_category', 
            y='incident_time',
            labels = {"incident_time_category":"Time when the incident occured", "incident_time":"Number of incidences" }, 
            title = "Distribution of incidences as per the time of its occurance",
            color='incident_time_category', 
            color_discrete_map={'Evening': 'red','Night': 'blue','Morning': 'green','Early Morning': 'orange'})

fig.show()

<Figure size 720x504 with 0 Axes>

Most of the incidents take place in the evening (between 12.01 to 18.00 hours) followed by night (between 18.01 to 24.00 hours), morning (between 06.01 to 12.00 hours) and early morning (between 00.00 to 06.00 hours).

df2 = df.groupby(['Incident Category','incident_subcategory']).count().reset_index()
df2.rename(columns = {'incident_id':'count'}, inplace = True)
df2

	Incident Category	incident_subcategory	incident_datetime	incident_date	incident_time	incident_year	incident_day_of_week	report_datetime	row_id	count	...	police_district	cad_number	intersection	cnn	analysis_neighborhood	supervisor_district	latitude	longitude	time_difference	incident_time_category
0	Arson	Arson	1920	1920	1920	1920	1920	1920	1920	1920	...	1920	1920	1920	1920	1920	1920	1920	1920	1920	1920
1	Assault	Aggravated Assault	15517	15517	15517	15517	15517	15517	15517	15517	...	15517	15517	15517	15517	15517	15517	15517	15517	15517	15517
2	Assault	Simple Assault	25402	25402	25402	25402	25402	25402	25402	25402	...	25402	25402	25402	25402	25402	25402	25402	25402	25402	25402
3	Burglary	Burglary - Commercial	4729	4729	4729	4729	4729	4729	4729	4729	...	4729	4729	4729	4729	4729	4729	4729	4729	4729	4729
4	Burglary	Burglary - Hot Prowl	4773	4773	4773	4773	4773	4773	4773	4773	...	4773	4773	4773	4773	4773	4773	4773	4773	4773	4773
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
92	Warrant	Other	10601	10601	10601	10601	10601	10601	10601	10601	...	10601	10601	10601	10601	10601	10601	10601	10601	10601	10601
93	Warrant	Warrant	9103	9103	9103	9103	9103	9103	9103	9103	...	9103	9103	9103	9103	9103	9103	9103	9103	9103	9103
94	Weapons Carrying Etc	Weapons Offense	3849	3849	3849	3849	3849	3849	3849	3849	...	3849	3849	3849	3849	3849	3849	3849	3849	3849	3849
95	Weapons Offense	Weapons Offense	4582	4582	4582	4582	4582	4582	4582	4582	...	4582	4582	4582	4582	4582	4582	4582	4582	4582	4582
96	nan	nan	559	559	559	559	559	559	559	559	...	559	559	559	559	559	559	559	559	559	559

97 rows × 27 columns

Heat Map to understand which category has the most number of crimes

fig = px.treemap(df2, path=[px.Constant("all"), 'Incident Category', 'incident_subcategory'], values='count')
fig.update_traces(root_color="lightgrey")
fig.update_layout(margin = dict(t=50, l=25, r=25, b=25))
fig.show()

Incident Hotspots on the map of San Francisco

fig, ax = plt.subplots(figsize=(35,35))
ax = plt.axes(projection=crs.PlateCarree())
ax.coastlines()
ax.set_extent([-122.6, -122.3, 37.68, 37.85])
ax.gridlines(draw_labels=True)

plt.scatter(x=df.longitude, y=df.latitude, 
            color="red", s=20, transform=crs.PlateCarree())

plt.show()

# Get a count of the crimes by neighborhood, returns a series
crime_neighbourhood = df.police_district.value_counts()
# Stick it into a data frame.
neighbour_df = pd.DataFrame(crime_neighbourhood)

# Set up the data frame with the required structure.
neighbour_df.index.name = 'neighborhood'
neighbour_df.reset_index(inplace=True)
neighbour_df.rename(columns={'police_district': 'No of Crimes'}, inplace=True)
neighbour_df

	neighborhood	No of Crimes
0	Central	100613
1	Northern	91518
2	Mission	84718
3	Southern	81150
4	Tenderloin	64532
5	Bayview	59492
6	Ingleside	50833
7	Taraval	47267
8	Richmond	41986
9	Park	31274
10	Out of SF	19815

San Francisco Map using Folium

# San Francisco latitude and longitude values
lat = 37.77
long = -122.42

# Create map using folium
sf1_map = folium.Map(location=[lat, long], zoom_start=12)

# display the map of San Francisco
sf1_map

Make this Notebook Trusted to load map: File -> Trust Notebook

Plotting Neighbourhoods on San Francisco Map

# Read in the geojson file with the neighborhoods and lat/long 
#sf_geo  = r'SanFrancisco.Neighborhoods.json'
sf_geo = r"https://cocl.us/sanfran_geojson"

# Create the map, centered on San Francisco with zoom level 12. 
SF_map = folium.Map(location=[37.7749, -122.4194], zoom_start=12)

# Add the choropleth to the map, with the wanted options.
SF_map.choropleth(
    geo_data=sf_geo,
    name='Choropleth',
    data = neighbour_df,
    columns=['neighborhood','No of Crimes'], 
    key_on='feature.properties.DISTRICT', 
    fill_color = 'YlOrRd',
    fill_opacity = 0.7,
    line_opacity = 0.2,
    legend_name = 'Crimes in SF'
)

# creating a state indexed version of the dataframe so we can lookup values
# neighbor_index = neighbour_df.set_index('neighborhood')
  
# looping thru the geojson object and adding a new property(unemployment)
# and assigning a value from our dataframe
# for s in cp.geojson.data['features']:
#     s['properties']['unemployment'] = state_data_indexed.loc[s['id'], 'Unemployment']
  
#   # and finally adding a tooltip/hover to the choropleth's geojson
# folium.GeoJsonTooltip(['name', 'unemployment']).add_to(cp.geojson)
  
# folium.LayerControl().add_to(m)

# folium.LayerControl().add_to(SF_map)
SF_map

/opt/conda/lib/python3.8/site-packages/folium/folium.py:407: FutureWarning:

The choropleth  method has been deprecated. Instead use the new Choropleth class, which has the same arguments. See the example notebook 'GeoJSON_and_choropleth' for how to do this.

Make this Notebook Trusted to load map: File -> Trust Notebook

References

• https://stackoverflow.com/questions/70471888/text-as-tooltip-popup-or-labels-in-folium-choropleth-geojson-polygons

• https://stackoverflow.com/questions/46223224/matplotlib-plot-countplot-for-two-or-more-column-on-single-plot

• https://matplotlib.org/stable/tutorials/introductory/customizing.html

• https://medium.com/@kvnamiparaa-better-visualisation-of-pie-charts-by-matplotlib-935b7667d77f

• https://scitools.org.uk/cartopy/docs/latest/matplotlib/feature_interface.html

• https://plotly.com/python/bar-charts/